How reliable are Arduino Uno boards?

Then you can sell the system monitoring as a service.

Look at the Allen-Bradley MicroLogix or Micro800 series PLCs, with the PanelView800 HMIs. Other companies make similar products, but Allen-Bradley/Rockwell Automation is the biggest US manufacturer. They're more expensive than an Arduino, but they'll actually last in the environment and will be easier to program if you know what you're doing. They'll also make it possible for you to actually get all of the approvals/listings you'll need.

Arduinos are not suitable for industrial use. You as a manufacturer will likely be required to comply with certain standards which arduino won't be likely to comply with. There is no environmental and EM shielding on those boards, connectors on them aren't reliable either. And there is also software, I wouldn't trust arduino library with any critical task. Look in to industrial PLCs, or design your own.

What happens if your system goes haywire during an extensive operation. Suddenly temperatures start rising, palms become sweaty, doctors spill their spaghetti.
In all seriousness though, I assume there are standards in place that you need to conform to.
Arduino is good to prototype with, and for non-critical applications maybe.

tbh i didn't even know those existed, this isn't my field, i guess this would make much job much easier.
i won't have to build a shield box, buttons, etc... and i won't have to worry about reliability, just programming.
but i guess those things aren't cheap and would eat on my profit margin.
do you have a price range?

fair enough, i guess you're right.
thanks.

wait, the Micrologix cost $320 ! thats 10 times more expensive than an Arduino, fuck that

Why the fuck would you use an Arduino in production? The Uno is just an ATmega328 or ATmega328P with some other shit on it like a crystal and a poor man's SPI-based AVR programmer.

THIS.

Medical stuff uses PLCs not microprocessors for reliability reasons.

An arduino based system would never get certified.

What do you think is inside a PLC?
The difference between a plc and a Arduino are, hardware wise, in things like
- EM-Shielding
- Protection against the environment
- Isolated inputs (galvanic)
- Better quality assurance

Programming plc's is shit in my opinion though. Something more robust than an Arduino would be nice.
But we really need to know more about the requirements of the system both functional and with regards to stuff like certifications.

How about controllino.biz/product/
Basically arduino in PLC format.
For surgery rooms I think you will need to comply with
Read more at electronicdesign.com/power/know-your-regulations-you-design-medical-electronics I'm not sure how much of this applies to AC devices since I'm not sure if AC is considered a medical device, but there should be standard somewhere that states requirements for AC as well.
There are safety standards for software as well so I hope you enjoy testing. Medical equipment is designed to a lot and very rigorous standards, I don't think surgeons will want AC to fail when they are in the middle of the surgery.

the controlino mini seems noice, but still feels horribly over priced.
a couple of relays and shielding for $100 more, that doesn't seem right.
i'm fairly confident i could build something with exactly the same spec for half that

I bet that the half of the cost is in housing. Injection molding isn't cheap if you don't produce a lot of housings. But then again why does the price of the electronics have to be so low? Aren't medical devices very expensive by default? If I was building such device I would go with PLC from some reputable manufacturer such as Siemens or similar. Rolling your own is also a good option if you follow all prescribed standards, so if you have knowledge on how to design PCBs, by all means go with custom solution.

damn dude, you sound like a coorporate curry monkey that can only think with acronyms.

Not the same guy, but in EE things are done differently than in software world, it's more "corporate", but damages caused by improper implementation are more severe as well. If a web app breaks no one will die, but you can't say the same about flight computer or ECU in your car. Those things need to work all the time and are not allowed to crash.

idk m8, gonna have to check with the owner again.
i also think a PLC would be the way to go, but it all depends on how much he's willing to pay for it.

i was aiming for a $200 to $300 price range with an Arduino, but with the PLC it would be around $700.
i should check with him.

i'm not really familiar with PLCs, do you have any particular model to recommend?
the Siemens LOGO seem to fit

Choice of PLC depends entirely on requirements of the application e.g. inputs/outputs and complexity of logic. I'm not familiar with requirements for AC units installed in medical facilities but for LOGO is stated that it has certification for installation in residential areas so it will be very reliable as long as you don't expose it to extreme industrial environments, for instance you would use it to control traffic lights. You will need to check if there are any regulations concerning standards on the devices you are trying to sell. Last thing you want is buying some controller only to find out that you cant use it because it was not approved for use in medical institutions.

you can, I've seen companies that do but they still do the shielding.

nothing complex, even an Arduino seems over kill.
just need to check the temp using a temp probe and start the compressor accordingly.
3 other probes are there to check if everything is working fine.
thats it.

there are 4 outputs max
and 4 inputs.

not counting the lcd display and the buttons

Well in that case all you need to find out is what safety regulations you need to comply with, that's the main issue that needs to be addressed here. If there aren't any then you could go with arduino or bare microcontroller or even all discrete components and make it as cheap as possible. But if there are regulations you will need to add all required modifications to your board, which when considering development time you total project cost will exceed cost of just buying a PLC.

yeah this is true if it's one object don't forget your opportunity cost modifying it and the fact that you may need a medical certified pws(there's a proper term for this but I don't have it sorry), don't cheap out because the regulations hurt if you fuck up

Don't forget OP that to interface with a PLC you will also need components that can do so. RTD's to measure temperature? You'd better be an expert at programming PI loops and be ready to shell out for an analog input card. Motor contactors? If you want something reliable, $$$.

PLCs are the right way to do what you're trying to do. There may be some CompactLogix with Analog inputs that would be suitable, but it would still be way out of your budget.

I have no idea how you're being contracted to do this when you have so little understanding of the surrounding necessary knowledge to do a half-decent job.

On-off control of a compressor should not even be remotely tolerated by any intelligent client as it will be noisy, incredibly distracting, and wear out the equipment, prematurely. VFD driven would be the only way to go, but again, $$$.

t. industrial PLC maintenance electrician

What is that "t. blah blah" shit?

Where did this habit of ending messages come from, and what does "t." abbreviate?

Sauce on grill?

>>>Holla Forums

it's finnish
kinda like "sincerely, bla bla bla"

this
t. feels goob beingi a finlandese man

reliable enough. but there may be other issues--

Firstly... in some countries the equipment used in a medical environment does need to pass special certifications. Finding out if this is true for you would be the first order of business here. The government regulations may make this a job way out of your league (no offense intended).

Secondly,,, PLCs aren't anything special; an Arduino with a bunch of relays connected could do the exact same things. The main reason for PLC's existence is so janitors can alter the 'program' without knowing dick about computers. This might be an instance where that would never be necessary. In OP's description the arduino isn't really "controlling" anything, it is just monitoring the state of other components that exist and signaling if they appear to be malfunctioning.

If you did this, it would be better to use the Uno CPU and just build a "minimal Arduino" off that. This way, all of the components on the board can be separately replaceable and it still doesn't cost much to do. Search google for "minimal arduino" to see how easy that is. The Arduino Uno processor needs very few parts to be functional.

(girl from)
.....that likely means a girl that works at a Starbucks in Portland, and who's never visited and doesn't know shit about the USSR

No man you don't understand because the USSR wasn't TRUE Communism blablabla No True Scottsman blablabla something something more logical fallacies bla bla bla private property didn't always exist something something retardation.

...

You are 100% not fit for this job.

Hey man, I don't give a fuck if she's a delusional commie, that's some choice puffy vulva.

You don't stick your dick in crazy. She probably also has AIDS, so now you fucked a crazy commie bitch and now you have AIDS. Good job.

Awesome, local sauce is best sauce

could someone recommend a good PLC for the job?

The first thing you need to do is figure out the control scheme. Ask him how he has controlled these in the past, what problems he's run into, etc. To get a feel for the job.

Next is identifying what exactly it is that you want to monitor; these are called "process variables" that will be measured with things like temperature probes, flow meters, pressure gauges, etc. Try to decide how much information you actually need to take in to determine alarming.

For example, you cannot assume that just because your compressor is on, that cold fresh air is actually getting into the room - there could be a clogged air filter, there could be a leak in the coolant, etc. You would need to know the rate and temperature of air flow at the exhaust vent to be sure. If the flow is below a certain point, but the temperature is fine, could be a clogged air filter, if it's hot air but lots of flow, probably a coolant leak. If it's neither cold nor has flow but your output to your compressor is on, fan might be bad or you may have a blown fuse for your fan power.

So, start selecting components and see how they interface. If there are reliable USB digital thermometers, flow meters, etc that have acceptable accuracy, arduino may be feasible. Or if arduinos have analog input cards, your choices become even greater.

All of the electrical engineers in charge of the most massive industries and economies of scale that the world has ever seen say that guy should stick to commenting on subjects with which he has familiarity.
There are cases where you could use a sprinkler timer instead of a plc very successfully, there are cases where an arduino would do nicely, and there cases where nothing else is appropriate. Memes have no place in the fucking operating room where your carelessness could quite literally kill someone.